Abstract: The invention relates to a computer-implemented method for programming a checkpoint of at least one application executed on a computing infrastructure. The computing infrastructure includes a plurality of resources allocated to the execution of the application. The method includes determining a date of failure of the application as a function of a prediction of an anomaly of at least one resource of the plurality of resources, and programming the checkpoint as a function of the date of failure.

Abstract: An aspect of the invention relates to a method for a plurality of clients to access a file in a distributed file system, the file being replicated on at least one other server, the method comprising the steps of: Receiving, by an access management device, a request to access the part of the file sent by a first client from the plurality of clients, Selecting a first server based on at least one parameter, Authorization the first client to access the file stored by the first server Receiving, following authorization, a request to access the same file sent by a second client, the second client being different from the first client, Selecting a second server based on the parameter, the second server being different from the first server, Authorizing the second client to access the file stored by the second server.

Abstract: One aspect of the invention relates to a method for optimizing the energy consumption of a computing infrastructure comprising a plurality of compute nodes, each associated with a plurality of jobs, the method comprising the following steps: Receiving an energy consumption reduction request comprising an energy consumption reduction objective; Determining a set of jobs to be suspended, the set of jobs being determined based on the received energy consumption reduction objective; Sending to a runtime environment of each compute node associated with a job comprised in the determined set of jobs, a suspension request comprising said job; Upon receiving a suspension request, suspending the job comprised in the received suspension request, the suspension being carried out via a set of non-native instructions implemented in the runtime environment of each compute node; Active waiting by the suspended job for a suspension stop message comprising said suspended job, the active waiting being carried out via a set o

Abstract: A method for managing message delivery in a computing infrastructure. For each message to be delivered, the message is sent simultaneously by a transmitting component, to each server of a plurality of receiving servers. For each server that received the sent message, the server computes a signature specific to the message received, which is identical for each server, and the server sends the computed signature to a synchronization component. The sent signature is received by the synchronization component. If a first condition according to which the received signature is not stored in a signature database is met, the received signature is stored in the signature database and an instruction is sent to store or transmit the received message to the server by the synchronization component. Otherwise, the synchronization component sends an instruction to the server to delete the received message and the computed signature.

Abstract: An aspect of the invention relates to a method of managing jobs in a information system (SI) on which a plurality of jobs run, the information system (SI) comprising a plurality of computer nodes (NDi) and at least a first storage tier (NS1) associated with a first performance tier and a second storage tier (NS2) associated with a second performance tier lower than the first performance tier, each job being associated with a priority level determined from a set of parameters comprising the node or nodes (NDi) on which the job is to be executed, the method comprising a step of scheduling the jobs as a function of the priority level associated with each job; the set of parameters used for determining the priority level also comprising a first parameter relating to the storage tier to be used for the data necessary for the execution of the job in question and a second parameter relating to the position of the data necessary for the execution of the job (TAi) in question.

Abstract: A high-performance computing device with adaptable computing power in which several high-performance computers exist that are isolated from each other to ensure data security. Each computer is run by a local resource manager that manages the computer at the node level. All or some of the local managers are managed via a global resource manager that the high-performance computing device manages at the level of the computers it contains.

Abstract: A method of aiding the design of a data center is disclosed. In one aspect, the method refers to generating a list of components from a library that satisfy a defined list of needs.

Abstract: The invention relates to a method for managing a buffer memory space associated with a persistent data storage system of a computing machine. The buffer memory space is suitable for temporarily storing in the RAM of the machine one or more portions of a single data file of the persistent data storage system that was previously accessed by one or more processes executed on the machine. The operating system of the machine tracks each of the portions of the file that are projected in the buffer memory space by a descriptor belonging to a plurality of buffer memory projection descriptors which are all associated with the tracking of one or more portions of the file projected in the buffer memory space.

Abstract: An aspect of the invention relates to a method for a plurality of clients to access a file in a distributed file system, the file being replicated on at least one other server, the method comprising the steps of: Receiving, by an access management device, a request to access the part of the file sent by a first client from the plurality of clients, Selecting a first server based on at least one parameter, Authorization the first client to access the file stored by the first server Receiving, following authorization, a request to access the same file sent by a second client, the second client being different from the first client, Selecting a second server based on the parameter, the second server being different from the first server, Authorizing the second client to access the file stored by the second server.

Abstract: The invention relates to a method for managing a buffer memory space associated with a persistent data storage system of a computing machine. The buffer memory space is suitable for temporarily storing in the RAM of the machine one or more portions of a single data file of the persistent data storage system that was previously accessed by one or more processes executed on the machine. The operating system of the machine tracks each of the portions of the file that are projected in the buffer memory space by a descriptor belonging to a plurality of buffer memory projection descriptors which are all associated with the tracking of one or more portions of the file projected in the buffer memory space.

Abstract: Processes pending execution in respective compute nodes (N1-Nn) of a cluster of a distributed HPC computer can communicate with one another by message exchange through an interconnected fabric. To exchange messages between the processes a method is proposed to identify the physical cards associated with the compute nodes directly from the hostname of the compute nodes as they are used in the user program. This direct identification is made from at least one mapping table (MT1) associating bijectively the hostname of each compute node of the cluster with the unique logical address of the associated physical card. This mapping table is kept in a component of the computer responsible for managing the fabric, i.e. the fabric manager (FM1). Various implementations enable scaling when accomplishing the method of message exchange between processes.

Abstract: The saving of data in an information technology (IT) infrastructure offering activity resumption functions is disclosed. For these purposes, a saving system is provided with at least one first and one second sets of data storage and at least one first and one second memory controllers associated with the first and second data storage, respectively. The saving system is furthermore provided with a microcontroller configured to duplicate a stream of commands and of data destined for the first memory controller to the second memory controller in a mode of normal use allowing a local replication of data in the second set of data storage and to address a distinct stream of commands and of data to each of the first and second memory controllers in a mode of remote replication.

Abstract: A screen copying method including the copying, in the form of an image file, of a portion of an image displayed on a screen where a user graphically selects the image portion, the copying of the selected image portion automatically resulting in: the copying of the pointer or pointers associated with the image portion along with the coordinates of the pointers in the image portion, the pointers respectively pointing to objects; and the maintaining of the respective links between the pointers and the coordinates so that, during a subsequent redisplay of the image portion on a screen, the graphical selection of a point corresponding to one of the coordinates by a user automatically activates the pointer corresponding to the coordinate of the selected point and brings up the object corresponding to the activated pointer.

Abstract: A method of remote monitoring and remote control of a cluster including nodes connected to a communication network of a first type, a relay node of the nodes including first and second network interfaces in accordance with first and second types of communication network respectively, the first and second types being distinct, some steps of the monitoring and control method being implemented in a remote computer linked to the relay node by a communication network of the second type, the method including receiving a packet via the first network interface; encapsulating the received data packet in a data frame in accordance with a protocol of the communication network of the second type; sending the data frame to the remote computer via the second network interface, the receiving of the data packet, the encapsulating of the data packet and the sending of the data frame being implemented in the relay node.

Abstract: For the management of a file system for accessing data in a storage system in which the data are stored physically in a unique manner, a first storage environment associated with a first access performance level is mounted, from a first mount point. Moreover, at least one second storage environment, different from the first storage environment, and associated with a second access performance level is mounted, from a second mount point and with total or partial overlay of data with respect to the first storage environment. The data physically stored in a unique manner in the storage system is accessed, either via the first mount point or via the second mount point, as a function of a data use case.

Abstract: A method for automatically managing the electricity consumption of a server farm including a plurality of nodes, the method including measuring an instantaneous consumption of the server farm; acquiring an instantaneous consumption limit; predicting a future consumption according to a function of at least the instantaneous consumption measurement; and, when the prediction is higher than the acquired instantaneous limit, selecting at least one node and electrically switching off the at least one selected node.

Abstract: A device for indicating a rack among a plurality of racks, the rack being configured to receive a plurality of pieces of computer equipment is disclosed. In one aspect, the device comprises a communication unit configured to receive at least one signal from at least one piece of equipment of the pieces of equipment. The signal comprises information enabling a state of the piece of equipment to be determined. The device further comprises a control unit configured to determine a state of the rack based at least in part on the signal. The device further comprises a display unit for displaying a representation of the state determined by the control unit.

Abstract: A method for monitoring the operation of an IT infrastructure including a plurality of calculation nodes, includes selecting calculation nodes for performing a calculation, performing the calculation via the selected calculation nodes, attributing, via the sequencer, a score to each one of the calculation nodes having participated in the calculation performed, with each score reflecting a difference between a measured operating parameter of the calculation node for which the score is attributed and a reference operating parameter of the calculation node for which the score is attributed, verifying the operation of the calculation nodes having participated in the calculation performed, the verification being carried out using scores attributed to the calculation nodes having participated in the calculation.

Abstract: The method relates to the monitoring of at least one routing parameter for a cluster including nodes and switches, static communication links connecting nodes and switches. Each switch includes several output ports. After having selected at least one switch, a number of routes per port is calculated for each port of each switch selected, routes being defined during a routing step for each connecting one node to another. A mean number of routes per port is then calculated for the at least one selected switch. Each number of routes per port calculated is then compared with the mean number of routes per port calculated and, in response to this comparison, a potential imbalance of routing of the cluster is notified.

Abstract: Processes pending execution in respective compute nodes (N1-Nn) of a cluster of a distributed HPC computer can communicate with one another by message exchange through an interconnected fabric. To exchange messages between the processes a method is proposed to identify the physical cards associated with the compute nodes directly from the hostname of the compute nodes as they are used in the user program. This direct identification is made from at least one mapping table (MT1) associating bijectively the hostname of each compute node of the cluster with the unique logical address of the associated physical card. This mapping table is kept in a component of the computer responsible for managing the fabric, i.e. the fabric manager (FM1). Various implementations enable scaling when accomplishing the method of message exchange between processes.